Learning to Recognize Talkers from Natural, Sinewave, and Reversed Speech Samples

Size: px
Start display at page:

Download "Learning to Recognize Talkers from Natural, Sinewave, and Reversed Speech Samples"

Transcription

1 Learning to Recognize Talkers from Natural, Sinewave, and Reversed Speech Samples Presented by: Pankaj Rajan Graduate Student, Department of Computer Sciences. Texas A&M University, College Station

2 Agenda Introduction Background Experiments Discussion

3 What the paper is about? How a listener recognizes the unfamiliar talker? Different talker specific properties of the speech signal to perceptual learning

4 Traditional View: Word and Talker Recognition Features for Speech perception and Talker recognition are different. As per Traditional View: : Vocal quality of Individual is represented by Features that are linguistically irrelevant. Melodic patterns Roughness Smoothness Fundamental Frequency Glottal Features Style/Accent For perception talker specific attributes are separated from phonetic representation.

5 Alternative to Traditional view Experiments have revealed relation between Linguistic and Indexical perception. Evidence!!! Experiments show that Listeners recognize phonemes and words faster and accurate when spoken by same speaker. Talker attributes cannot be ignored. Linguistic perception preserve talker specific attributes which become part of long term memory. Experiments by Nygaard In which he found that the listeners familiar with the talker were able to transcribe the sentences more easily

6 Conclusions of Alternative view Talker properties and the linguistic form are associated with same single memory system. Linguistic and Indexical attributes are transmitted in parallel

7 Remez Experiment AIM: To test hypothesis that phonetic representations represent both talker characteristics and words. Listener familiar with the talker were presented the Sinewave representation of the natural speech for individual recognition. Listeners were able to recognize the talker and hence demonstrated that perception of talker characteristics is present in talker specific phoneme utterances. Sine Waves: Acoustic structure of the original utterance that lack fine grained acoustic details of natural speech but evoke: 1. impression of segmental phonetic attributes and 2. Vocal quality like Fundamental frequency, broadband resonances, harmonic structure.

8 Spectrogram Representation

9 Present Work To establish if learners can perceive speaker in absence of acoustic vocal quality parameters. To know about the feature structure of the talker that develops during perceptual training To know the extent to which the attributes of the talker are transferred to the type of signal.

10 5 Experiments: Experiment 1 and 2 Aimed to investigate if listener would recognize a talker from samples featuring phonetic segmental properties at the cost of qualitative aspects. Experiment 3, 4,5 Experiment 3, 4,5 Aimed to explore how listeners would categorize unfamiliar talker when presented with qualitative aspects at the cost of segmental phonetic and lexical properties

11 Experiment 1 and 2

12 Experiments: Both Sinewave and the natural speech samples were used to train the listeners. Sentences were recorded in an audiotape in a soundproof booth and were low passed filtered at 4.5 khz and sampled at 10 khz. Subjects were trained to extent of 70 % accuracy to identify the talker and training was conducted for several days. Subjects were encouraged to pay attention to the talker attributes rather than content of message.

13 Continued.. There was a familiarization phase, to reinstate correspondence between sine wave tokens and talkers name. It was followed by the Generalization test

14 Experiment1learning to identify talkers with Sinewave Sentences Training: Listeners were trained in a quite room to identify 10 speakers of Sinewave utterances Analysis: Listeners were able to identify speakers and accuracy increases day by day. One way ANOVA revealed significant effect of talker identity on recognition performance Generalization Performance/Test: Half of the subjects were first presented with natural speeches to be identified before Sinewave and other half was presented in reversed order. Results: Talker specific knowledge acquired during Sinewave training generalized to novel natural and Sinewave sentences

15 Continued Listener can identify talker with the phonetic attributes present in the Sinewave in absence of traditional qualitative attributes of vocal sound production. It is perceptual discriminability of talker in the set that is source of differences in identification. Interesting Observation Individual listener differed in their ability to identify the talkers. However, reason was not known. Conclusion: Individual attributes are carried by the segmental phonetic properties in addition to vocal timber. Linguistic and individual attributes can be represented by common representation code.

16 Training

17 Last day of Training Experiment 1

18 Generalized test

19 Experiment2Learning to identify talkers from natural speech Aim: To find if similar generalization as observed in previous experiment can be obtained using natural speech training Training: Subjects were trained on the natural sentences Similar to training in the Experiment 1. Analysis Listeners were able to learn from natural speech very fast. It indicates that natural speech provides listeners with salient sample of each talker s indexical attributes.

20 Continued. Generalized performance: Similar to the first Experiment. Results: Listener s ability to understand the speech was 88% for natural speech generalization test but 27% for Sinewave generalization test. Conclusions Indexical knowledge acquired during the training with the natural speech doesn't generalized to Sinewave utterances.

21 Last day Training Results- Exp:2

22 Generalization Test

23 Experiment 3, 4, 5

24 Experiment 3 Recognizing an unintelligible talker AIM: To test hypothesis that subjects use glottal source quality rather than fine grained phonetic properties during natural speech training. Training: Listeners were trained using natural speech. Generalization Performance: Generalization test samples were reversed natural speech and sinewave samples. Analysis: Subjects were able to identify talker from reversed speech samples Performance on the Sinewave samples was far poorer

25 Continued.. Results Qualitative attributes are prominent perceptual attributes during training with natural speech samples. Though results were poorer on Sinewave samples, correlation analysis shows that listeners do encoded some fine grained phonetic attributes for the recognition. Conclusion Under Normal circumstances listeners naturally encode talkers using mixture of different properties, relying more on qualitative characteristics while also using the phonetic details

26 Generalization test: Exp:3

27 Experiment 4 Getting to know unintelligible talker Aim: To test the sufficiency of qualitative aspects of speech in relative absence of many phonetic and lexical impressions. Ability to learn from reversed samples is tested. Training Training material were reversed speech samples. Generalization Process The generalized material are reversed speech samples and sine wave speeches sentences. Analysis Listener learned to identify the individuals from reversed speech samples. Rate was intermediate between training with natural speech and with sine wave replica.

28 Continued.. Results Reveres speech training was highly correlated with reverse speech talker identification. No correlation between training and Sinewave talker identification. Conclusion The features with which listeners become familiar during training did not correspond to indexical attributes available in Sinewave replica.

29 Last day Training Performance

30 Generalization test Exp:4

31 Experiment5 Robustness of Qualitative Indexical attributes Aim Training To test that the qualitative attributes available in reversed speech samples match those of natural speech. Reversed natural speech samples. Generalized performance Analysis Results Generalized test material were natural speech samples and Sinewave replicas. Results were identical with the previous experiment. Similar to the previous experiment. Weak Correlation between reversed speech training and Sinewave generalization.

32 Generalization test Exp:5

33 Discussion: The listeners exploit various cues to perceive the talker. In some circumstances global qualitative attributes are used to perceive talker while in some circumstances, phonetic attributes alone can be used.

RESEARCH ON SPOKEN LANGUAGE PROCESSING Progress Report No. 24 (2000) Indiana University

RESEARCH ON SPOKEN LANGUAGE PROCESSING Progress Report No. 24 (2000) Indiana University SPEECH AND IMPLICIT MEMORY RESEARCH ON SPOKEN LANGUAGE PROCESSING Progress Report No. 24 (2000) Indiana University Speech Perception and Implicit Memory: Evidence for Detailed Episodic Encoding of Phonetic

More information

Stress and Accent in Tunisian Arabic

Stress and Accent in Tunisian Arabic Stress and Accent in Tunisian Arabic By Nadia Bouchhioua University of Carthage, Tunis Outline of the Presentation 1. Rationale for the study 2. Defining stress and accent 3. Parameters Explored 4. Methodology

More information

Relationship Between Acoustic Measures and Judgements of Intelligibility in Parkinson s Disease: A Within-Speaker Approach

Relationship Between Acoustic Measures and Judgements of Intelligibility in Parkinson s Disease: A Within-Speaker Approach Relationship Between Acoustic Measures and Judgements of Intelligibility in Parkinson s Disease: A Within-Speaker Approach Lynda Feenaughty, Kris Tjaden, & Joan Sussman Clinical Linguistics and Phonetics,

More information

Speech Signal Processing: An Overview

Speech Signal Processing: An Overview Speech Signal Processing: An Overview S. R. M. Prasanna Department of Electronics and Electrical Engineering Indian Institute of Technology Guwahati December, 2012 Prasanna (EMST Lab, EEE, IITG) Speech

More information

Lecture 1-8: Audio Recording Systems

Lecture 1-8: Audio Recording Systems Lecture 1-8: Audio Recording Systems Overview 1. Why do we need to record speech? We need audio recordings of speech for a number of reasons: for off-line analysis, so that we can listen to and transcribe

More information

Signal Processing for Speech Recognition

Signal Processing for Speech Recognition Signal Processing for Speech Recognition Once a signal has been sampled, we have huge amounts of data, often 20,000 16 bit numbers a second! We need to find ways to concisely capture the properties of

More information

EVALUATION OF KANNADA TEXT-TO-SPEECH [KTTS] SYSTEM

EVALUATION OF KANNADA TEXT-TO-SPEECH [KTTS] SYSTEM Volume 2, Issue 1, January 2012 ISSN: 2277 128X International Journal of Advanced Research in Computer Science and Software Engineering Research Paper Available online at: EVALUATION OF KANNADA TEXT-TO-SPEECH

More information

L2 EXPERIENCE MODULATES LEARNERS USE OF CUES IN THE PERCEPTION OF L3 TONES

L2 EXPERIENCE MODULATES LEARNERS USE OF CUES IN THE PERCEPTION OF L3 TONES L2 EXPERIENCE MODULATES LEARNERS USE OF CUES IN THE PERCEPTION OF L3 TONES Zhen Qin, Allard Jongman Department of Linguistics, University of Kansas, United States qinzhenquentin2@ku.edu, ajongman@ku.edu

More information

The relation between the vocal tract and recorder sound quality.

The relation between the vocal tract and recorder sound quality. The relation between the vocal tract and recorder sound quality. Report on the tests carried out in January 1998 at the School of Physics, University of New South Wales in collaboration with professor

More information

PERCEPTION OF VOWEL QUANTITY BY ENGLISH LEARNERS

PERCEPTION OF VOWEL QUANTITY BY ENGLISH LEARNERS PERCEPTION OF VOWEL QUANTITY BY ENGLISH LEARNERS OF CZECH AND NATIVE LISTENERS Kateřina Chládková Václav Jonáš Podlipský Karel Janíček Eva Boudová 1 INTRODUCTION Vowels in both English and Czech are realized

More information

Listening Speaking Reading Writing

Listening Speaking Reading Writing Advanced CB 21 A One level Assess descriptions and narrations of factual and technical materials. Discriminate for accurate information while taking notes of a complex discourse. Assess the essential message

More information

How Robust Are Exemplar Effects in Word Comprehension?

How Robust Are Exemplar Effects in Word Comprehension? Running head: HOW ROBUST ARE EXEMPLAR EFFECTS? How Robust Are Exemplar Effects in Word Comprehension? Iris Hanique 1,2, Ellen Aalders 1, Mirjam Ernestus 1,2 1 Centre for Language Studies, Radboud University

More information

Turkish Radiology Dictation System

Turkish Radiology Dictation System Turkish Radiology Dictation System Ebru Arısoy, Levent M. Arslan Boaziçi University, Electrical and Electronic Engineering Department, 34342, Bebek, stanbul, Turkey arisoyeb@boun.edu.tr, arslanle@boun.edu.tr

More information

Rhythmic Perception, Music and Language: A New Theoretical Framework for Understanding and Remediating Specific Language Impairment

Rhythmic Perception, Music and Language: A New Theoretical Framework for Understanding and Remediating Specific Language Impairment 1 Rhythmic Perception, Music and Language: A New Theoretical Framework for Understanding and Remediating Specific Language Impairment Usha Goswami, Ruth Cumming and Angela Wilson Centre for Neuroscience

More information

On the perception of similarity among talkers

On the perception of similarity among talkers On the perception of similarity among talkers Robert E. Remez a Department of Psychology, Barnard College, 3009 Broadway, New York, New York 10027 Jennifer M. Fellowes Department of Psychiatry, New York

More information

Talker-specific Influences on Phonetic Boundaries and Internal Category Structure

Talker-specific Influences on Phonetic Boundaries and Internal Category Structure University of Connecticut DigitalCommons@UConn Master's Theses University of Connecticut Graduate School 6-13-2013 Talker-specific Influences on Phonetic Boundaries and Internal Category Structure Janice

More information

Crash course in acoustics

Crash course in acoustics Crash course in acoustics 1 Outline What is sound and how does it travel? Physical characteristics of simple sound waves Amplitude, period, frequency Complex periodic and aperiodic waves Source-filter

More information

Program curriculum for graduate studies in Speech and Music Communication

Program curriculum for graduate studies in Speech and Music Communication Program curriculum for graduate studies in Speech and Music Communication School of Computer Science and Communication, KTH (Translated version, November 2009) Common guidelines for graduate-level studies

More information

. Niparko, J. K. (2006). Speech Recognition at 1-Year Follow-Up in the Childhood

. Niparko, J. K. (2006). Speech Recognition at 1-Year Follow-Up in the Childhood Psychology 230: Research Methods Lab A Katie Berg, Brandon Geary, Gina Scharenbroch, Haley Schmidt, & Elizabeth Stevens Introduction: Overview: A training program, under the lead of Professor Jeremy Loebach,

More information

DRA2 Word Analysis. correlated to. Virginia Learning Standards Grade 1

DRA2 Word Analysis. correlated to. Virginia Learning Standards Grade 1 DRA2 Word Analysis correlated to Virginia Learning Standards Grade 1 Quickly identify and generate words that rhyme with given words. Quickly identify and generate words that begin with the same sound.

More information

SoundRecover2 the adaptive frequency compression algorithm More audibility of high frequency sounds

SoundRecover2 the adaptive frequency compression algorithm More audibility of high frequency sounds Phonak Insight April 2016 SoundRecover2 the adaptive frequency compression algorithm More audibility of high frequency sounds Phonak led the way in modern frequency lowering technology with the introduction

More information

RELATIONAL ACOUSTIC INVARIANCE IN JAPANESE VOWEL LENGTH DISTINCTION

RELATIONAL ACOUSTIC INVARIANCE IN JAPANESE VOWEL LENGTH DISTINCTION RELATIONAL ACOUSTIC INVARIANCE IN JAPANESE VOWEL LENGTH DISTINCTION Yukari Hirata Dept. of East Asian Languages and Literatures Colgate University, Hamilton, New York, USA yhirata@mail.colgate.edu ABSTRACT

More information

THE VOICE OF LOVE. Trisha Belanger, Caroline Menezes, Claire Barboa, Mofida Helo, Kimia Shirazifard

THE VOICE OF LOVE. Trisha Belanger, Caroline Menezes, Claire Barboa, Mofida Helo, Kimia Shirazifard THE VOICE OF LOVE Trisha Belanger, Caroline Menezes, Claire Barboa, Mofida Helo, Kimia Shirazifard University of Toledo, United States tbelanger@rockets.utoledo.edu, Caroline.Menezes@utoledo.edu, Claire.Barbao@rockets.utoledo.edu,

More information

Does elderly speech recognition in noise benefit from spectral and visual cues?

Does elderly speech recognition in noise benefit from spectral and visual cues? INTERSPEECH 14 Does elderly speech recognition in noise benefit from spectral and visual cues? Yatin Mahajan 1, Jeesun Kim 1, Chris Davis 1 1 The MARCS Institute, University of Western Sydney y.mahajan@uws.edu.au,j.kim@uws.edu.au,chris.davis@uws.edu.au

More information

Workshop Perceptual Effects of Filtering and Masking Introduction to Filtering and Masking

Workshop Perceptual Effects of Filtering and Masking Introduction to Filtering and Masking Workshop Perceptual Effects of Filtering and Masking Introduction to Filtering and Masking The perception and correct identification of speech sounds as phonemes depends on the listener extracting various

More information

Quarterly Progress and Status Report. Linguistic methods in communication aids

Quarterly Progress and Status Report. Linguistic methods in communication aids Dept. for Speech, Music and Hearing Quarterly Progress and Status Report Linguistic methods in communication aids Hunnicutt, S. journal: STL-QPSR volume: 29 number: 1 year: 1988 pages: 161-169 http://www.speech.kth.se/qpsr

More information

Formant Bandwidth and Resilience of Speech to Noise

Formant Bandwidth and Resilience of Speech to Noise Formant Bandwidth and Resilience of Speech to Noise Master Thesis Leny Vinceslas August 5, 211 Internship for the ATIAM Master s degree ENS - Laboratoire Psychologie de la Perception - Hearing Group Supervised

More information

DEPARTMENT: PSYCHOLOGY PREPARED BY: PSYCHOLOGY DEPT. DATE: FALL 2015

DEPARTMENT: PSYCHOLOGY PREPARED BY: PSYCHOLOGY DEPT. DATE: FALL 2015 FARMINGDALE STATE COLLEGE DEPARTMENT: PSYCHOLOGY PREPARED BY: PSYCHOLOGY DEPT. DATE: FALL 2015 COURSE TITLE: Cognitive Psychology COURSE CODE: PSY 272 CREDITS: 3 CONTACT HOURS: 45 CATALOG DESCRIPTION:

More information

IMPROVEMENT OF INTELLIGIBILITY PREDICTION OF SPOKEN WORD IN JAPANESE ACCENTED ENGLISH USING PHONETIC PRONUNCIATION DISTANCE AND WORD CONFUSABILITY

IMPROVEMENT OF INTELLIGIBILITY PREDICTION OF SPOKEN WORD IN JAPANESE ACCENTED ENGLISH USING PHONETIC PRONUNCIATION DISTANCE AND WORD CONFUSABILITY IMPROVEMENT OF INTELLIGIBILITY PREDICTION OF SPOKEN WORD IN JAPANESE ACCENTED ENGLISH USING PHONETIC PRONUNCIATION DISTANCE AND WORD CONFUSABILITY Teeraphon Pongkittiphan 1, Nobuaki Minematsu 1, Takehiko

More information

Quarterly Progress and Status Report. An ontogenetic study of infant speech perception

Quarterly Progress and Status Report. An ontogenetic study of infant speech perception Dept. for Speech, Music and Hearing Quarterly Progress and Status Report An ontogenetic study of infant speech perception Lacerda, F. and Aurelius, G. and Landberg,. and Roug-Hellichius, L-L. journal:

More information

Analog Representations of Sound

Analog Representations of Sound Analog Representations of Sound Magnified phonograph grooves, viewed from above: The shape of the grooves encodes the continuously varying audio signal. Analog to Digital Recording Chain ADC Microphone

More information

DEVELOPMENT AND ANALYSIS OF HINDI-URDU PARALLEL CORPUS

DEVELOPMENT AND ANALYSIS OF HINDI-URDU PARALLEL CORPUS DEVELOPMENT AND ANALYSIS OF HINDI-URDU PARALLEL CORPUS Mandeep Kaur GNE, Ludhiana Ludhiana, India Navdeep Kaur SLIET, Longowal Sangrur, India Abstract This paper deals with Development and Analysis of

More information

Quarterly Progress and Status Report

Quarterly Progress and Status Report Dept. for Speech, Music and Hearing Quarterly Progress and Status Report The relationship between residual hearing and speech intelligibility - Is there a measure that could predict a prelingually profoundly

More information

Bachelors of Science Program in Communication Disorders and Sciences:

Bachelors of Science Program in Communication Disorders and Sciences: Bachelors of Science Program in Communication Disorders and Sciences: Mission: The SIUC CDS program is committed to multiple complimentary missions. We provide support for, and align with, the university,

More information

Speech Emotion Recognition: Comparison of Speech Segmentation Approaches

Speech Emotion Recognition: Comparison of Speech Segmentation Approaches Speech Emotion Recognition: Comparison of Speech Segmentation Approaches Muharram Mansoorizadeh Electrical and Computer Engineering Department Tarbiat Modarres University Tehran, Iran mansoorm@modares.ac.ir

More information

Voice Quality. R. L. Starr Lecture for Language & Gender May 26, 2009

Voice Quality. R. L. Starr Lecture for Language & Gender May 26, 2009 Voice Quality R. L. Starr Lecture for Language & Gender May 26, 2009 Reminder: what aspects of human language can we study? high level low level What can we study? Discourse-level: how speakers interact

More information

Running head: IMPACT OF TALKER VARIATION

Running head: IMPACT OF TALKER VARIATION IMPACT OF TALKER VARIATION 1 Running head: IMPACT OF TALKER VARIATION Impact of Talker Variation in Following Context on Resolving Lexical Ambiguities in Speech Recognition Grace Park Department of Psychology,

More information

The Pronunciation of the Aspirated Consonants P, T, and K in English by Native Speakers of Spanish and French

The Pronunciation of the Aspirated Consonants P, T, and K in English by Native Speakers of Spanish and French 144 The Pronunciation of the Aspirated Consonants P, T, and K in English by Native Speakers of Spanish and French Philip Suarez Florida International University, USA Abstract: This study examines whether

More information

Perceptual experiments sir-skur-spur-stir

Perceptual experiments sir-skur-spur-stir Perceptual experiments sir-skur-spur-stir Amy Beeston & Guy Brown 19 May 21 1 Introduction 2 Experiment 1: cutoff Set up Results 3 Experiment 2: reverse Set up Results 4 Discussion Introduction introduction

More information

Intonation difficulties in non-native languages.

Intonation difficulties in non-native languages. Intonation difficulties in non-native languages. Irma Rusadze Akaki Tsereteli State University, Assistant Professor, Kutaisi, Georgia Sopio Kipiani Akaki Tsereteli State University, Assistant Professor,

More information

Objective Quality Assessment of Wideband Speech Coding using W-PESQ Measure and Artificial Voice

Objective Quality Assessment of Wideband Speech Coding using W-PESQ Measure and Artificial Voice Objective Quality Assessment of Wideband Speech Coding using W-PESQ Measure and Artificial Voice Nobuhiko Kitawaki, Kou Nagai, and Takeshi Yamada University of Tsukuba 1-1-1, Tennoudai, Tsukuba-shi, 305-8573

More information

Text-To-Speech Technologies for Mobile Telephony Services

Text-To-Speech Technologies for Mobile Telephony Services Text-To-Speech Technologies for Mobile Telephony Services Paulseph-John Farrugia Department of Computer Science and AI, University of Malta Abstract. Text-To-Speech (TTS) systems aim to transform arbitrary

More information

Text To Speech Conversion Using Different Speech Synthesis

Text To Speech Conversion Using Different Speech Synthesis INTERNATIONAL JOURNAL OF SCIENTIFIC & TECHNOLOGY RESEARCH VOLUME 4, ISSUE 7, JULY 25 ISSN 2277-866 Text To Conversion Using Different Synthesis Hay Mar Htun, Theingi Zin, Hla Myo Tun Abstract: Text to

More information

RESEARCH ON SPOKEN LANGUAGE PROCESSING Progress Report No. 22 (1998) Indiana University

RESEARCH ON SPOKEN LANGUAGE PROCESSING Progress Report No. 22 (1998) Indiana University THE PHONETICS OF PHONOLOGICAL SPEECH ERRORS RESEARCH ON SPOKEN LANGUAGE PROCESSING Progress Report No. 22 (1998) Indiana University The Phonetics of Phonological Speech Errors: An Acoustic Analysis of

More information

An Arabic Text-To-Speech System Based on Artificial Neural Networks

An Arabic Text-To-Speech System Based on Artificial Neural Networks Journal of Computer Science 5 (3): 207-213, 2009 ISSN 1549-3636 2009 Science Publications An Arabic Text-To-Speech System Based on Artificial Neural Networks Ghadeer Al-Said and Moussa Abdallah Department

More information

Aspects of Normal Language. COMD 326, Intro. 1

Aspects of Normal Language. COMD 326, Intro. 1 Aspects of Normal Language COMD 326, Intro. 1 Communication - exchanging information and ideas -encoding -transmitting (communication modes) -decoding COMD 326, Intro. 2 transmitting (communication modes)

More information

Only Prosody? Perception of speech segmentation in Kabyle and Hebrew

Only Prosody? Perception of speech segmentation in Kabyle and Hebrew Only Prosody? Perception of speech segmentation in Kabyle and Hebrew Amina Mettouchi 1, Anne Lacheret-Dujour 2, Vered Silber-Varod 4, Shlomo Izre el 3 1 Laboratoire de Linguistique de Nantes, Université

More information

Acoustic coustic Realization and Perception of English Lexical Stress by Mandarin Learners

Acoustic coustic Realization and Perception of English Lexical Stress by Mandarin Learners Acoustic coustic Realization and Perception of English Lexical Stress by Mandarin Learners Yuwen Lai University of British Columbia University of Kansas Joan Sereno University of Kansas Allard Jongman

More information

THE PHONETIC STUDY OF BILINGUALISM 1

THE PHONETIC STUDY OF BILINGUALISM 1 The Phonetic Study of Bilingualism 17 THE PHONETIC STUDY OF BILINGUALISM 1 James Emil Flege Individuals who learn a second language (L2) seldom manage to negotiate the full range of differences between

More information

On the Nature of Talker Variability Effects on Recall of Spoken Word Lists

On the Nature of Talker Variability Effects on Recall of Spoken Word Lists Journal of Experimental Psychology: Copyright 1991 by the American Psychological Association. Inc. Learning, Memory. and Cognition 278-7393/91/$3. 1991, Vol. 17, No. I, 152-162 On the Nature of Talker

More information

L9: Cepstral analysis

L9: Cepstral analysis L9: Cepstral analysis The cepstrum Homomorphic filtering The cepstrum and voicing/pitch detection Linear prediction cepstral coefficients Mel frequency cepstral coefficients This lecture is based on [Taylor,

More information

Lecture 12: An Overview of Speech Recognition

Lecture 12: An Overview of Speech Recognition Lecture : An Overview of peech Recognition. Introduction We can classify speech recognition tasks and systems along a set of dimensions that produce various tradeoffs in applicability and robustness. Isolated

More information

Data driven design of filter bank for speech recognition

Data driven design of filter bank for speech recognition Data driven design of filter bank for speech recognition Lukáš Burget 12 and Hynek Heřmanský 23 1 Oregon Graduate Institute, Anthropic Signal Processing Group, 2 NW Walker Rd., Beaverton, Oregon 976-8921,

More information

Using pronunciation data as a starting point in modeling word recognition

Using pronunciation data as a starting point in modeling word recognition Using pronunciation data as a starting point in modeling word recognition Mark A. Pitt and Keith Johnson Ohio State University, Columbus, OH E-mail: pitt.2@osu.edu, kjohnson@ling.ohio-state.edu ABSTRACT

More information

The effect of mismatched recording conditions on human and automatic speaker recognition in forensic applications

The effect of mismatched recording conditions on human and automatic speaker recognition in forensic applications Forensic Science International 146S (2004) S95 S99 www.elsevier.com/locate/forsciint The effect of mismatched recording conditions on human and automatic speaker recognition in forensic applications A.

More information

Plasticity in vowel perception and production: A study of accent change in young adults

Plasticity in vowel perception and production: A study of accent change in young adults Plasticity in vowel perception and production: A study of accent change in young adults Bronwen G. Evans and Paul Iverson Department of Phonetics and Linguistics, University College London, 4 Stephenson

More information

Literacy and prosody The case of low literate Senegalese learners of L2 Italian

Literacy and prosody The case of low literate Senegalese learners of L2 Italian Literacy and prosody The case of low literate Senegalese learners of L2 Italian Marta Maffia and Anna De Meo University of Naples L Orientale LESLLA Conference Nijmegen, 28 30 August 2014 Outline of the

More information

SPEAKER IDENTIFICATION FROM YOUTUBE OBTAINED DATA

SPEAKER IDENTIFICATION FROM YOUTUBE OBTAINED DATA SPEAKER IDENTIFICATION FROM YOUTUBE OBTAINED DATA Nitesh Kumar Chaudhary 1 and Shraddha Srivastav 2 1 Department of Electronics & Communication Engineering, LNMIIT, Jaipur, India 2 Bharti School Of Telecommunication,

More information

Lecture 1-10: Spectrograms

Lecture 1-10: Spectrograms Lecture 1-10: Spectrograms Overview 1. Spectra of dynamic signals: like many real world signals, speech changes in quality with time. But so far the only spectral analysis we have performed has assumed

More information

World Languages Portfolio Model. A new path to measuring growth in traditionally non-tested grades and subjects

World Languages Portfolio Model. A new path to measuring growth in traditionally non-tested grades and subjects World Languages Portfolio Model A new path to measuring growth in traditionally non-tested grades and subjects Benefits of the Model Authentic, individual growth scores for an additional group of teachers

More information

How Can Teachers Teach Listening?

How Can Teachers Teach Listening? 3 How Can Teachers Teach Listening? The research findings discussed in the previous chapter have several important implications for teachers. Although many aspects of the traditional listening classroom

More information

Things to remember when transcribing speech

Things to remember when transcribing speech Notes and discussion Things to remember when transcribing speech David Crystal University of Reading Until the day comes when this journal is available in an audio or video format, we shall have to rely

More information

PERCENTAGE ARTICULATION LOSS OF CONSONANTS IN THE ELEMENTARY SCHOOL CLASSROOMS

PERCENTAGE ARTICULATION LOSS OF CONSONANTS IN THE ELEMENTARY SCHOOL CLASSROOMS The 21 st International Congress on Sound and Vibration 13-17 July, 2014, Beijing/China PERCENTAGE ARTICULATION LOSS OF CONSONANTS IN THE ELEMENTARY SCHOOL CLASSROOMS Dan Wang, Nanjie Yan and Jianxin Peng*

More information

Perception of Flaps in American English and Korean

Perception of Flaps in American English and Korean Perception of Flaps in American English and Korean Eun-kyung Sung University of Delaware 1. Introduction Linguistic experience influences the mind of a listener. Listeners pick up the same acoustic signals

More information

This copy of the thesis has been supplied on condition that anyone who consults it is

This copy of the thesis has been supplied on condition that anyone who consults it is This copy of the thesis has been supplied on condition that anyone who consults it is understood to recognise that its copyright rests with its author and that no quotation from the thesis and no information

More information

HMM Speech Recognition. Words: Pronunciations and Language Models. Pronunciation dictionary. Out-of-vocabulary (OOV) rate.

HMM Speech Recognition. Words: Pronunciations and Language Models. Pronunciation dictionary. Out-of-vocabulary (OOV) rate. HMM Speech Recognition ords: Pronunciations and Language Models Recorded Speech Decoded Text (Transcription) Steve Renals Acoustic Features Acoustic Model Automatic Speech Recognition ASR Lecture 9 January-March

More information

The Roots of the Early Vocabulary in Infants Learning From Speech Daniel Swingley

The Roots of the Early Vocabulary in Infants Learning From Speech Daniel Swingley CURRENT DIRECTIONS IN PSYCHOLOGICAL SCIENCE The Roots of the Early Vocabulary in Infants Learning From Speech Daniel Swingley University of Pennsylvania ABSTRACT Psychologists have known for over 20 years

More information

SYNTHESISED SPEECH WITH UNIT SELECTION

SYNTHESISED SPEECH WITH UNIT SELECTION Institute of Phonetic Sciences, University of Amsterdam, Proceedings 24 (2001), 57-63. SYNTHESISED SPEECH WITH UNIT SELECTION Creating a restricted domain speech corpus for Dutch Betina Simonsen, Esther

More information

The Open University s repository of research publications and other research outputs

The Open University s repository of research publications and other research outputs Open Research Online The Open University s repository of research publications and other research outputs Stress and : Acoustic correlates of metrical prominence in Catalan Conference Item How to cite:

More information

62 Hearing Impaired MI-SG-FLD062-02

62 Hearing Impaired MI-SG-FLD062-02 62 Hearing Impaired MI-SG-FLD062-02 TABLE OF CONTENTS PART 1: General Information About the MTTC Program and Test Preparation OVERVIEW OF THE TESTING PROGRAM... 1-1 Contact Information Test Development

More information

Introducing Synthetic Phonics!

Introducing Synthetic Phonics! Introducing Synthetic Phonics! Synthetic phonics; it is a technical name and nothing to do with being artificial. The synthetic part refers to synthesizing or blending sounds to make a word. Phonics is

More information

Regionalized Text-to-Speech Systems: Persona Design and Application Scenarios

Regionalized Text-to-Speech Systems: Persona Design and Application Scenarios Regionalized Text-to-Speech Systems: Persona Design and Application Scenarios Michael Pucher, Gudrun Schuchmann, and Peter Fröhlich ftw., Telecommunications Research Center, Donau-City-Strasse 1, 1220

More information

Application Note. Using PESQ to Test a VoIP Network. Prepared by: Psytechnics Limited. 23 Museum Street Ipswich, Suffolk United Kingdom IP1 1HN

Application Note. Using PESQ to Test a VoIP Network. Prepared by: Psytechnics Limited. 23 Museum Street Ipswich, Suffolk United Kingdom IP1 1HN Using PESQ to Test a VoIP Network Application Note Prepared by: Psytechnics Limited 23 Museum Street Ipswich, Suffolk United Kingdom IP1 1HN t: +44 (0) 1473 261 800 f: +44 (0) 1473 261 880 e: info@psytechnics.com

More information

Front-End Signal Processing for Speech Recognition

Front-End Signal Processing for Speech Recognition Front-End Signal Processing for Speech Recognition MILAN RAMLJAK 1, MAJA STELLA 2, MATKO ŠARIĆ 2 1 Ericsson Nikola Tesla Poljička 39, HR-21 Split 2 FESB - University of Split R. Boškovića 32, HR-21 Split

More information

Practicing Pronunciation for Higher Scores in Standardized English Tests (TOEFL/IELTS)

Practicing Pronunciation for Higher Scores in Standardized English Tests (TOEFL/IELTS) SAUNDZ Practicing Pronunciation for Higher Scores in Standardized English Tests (TOEFL/IELTS) A White Paper Saundz Research 11/5/2014 The paper discusses pronunciation assessment methods on standardized

More information

Robust Methods for Automatic Transcription and Alignment of Speech Signals

Robust Methods for Automatic Transcription and Alignment of Speech Signals Robust Methods for Automatic Transcription and Alignment of Speech Signals Leif Grönqvist (lgr@msi.vxu.se) Course in Speech Recognition January 2. 2004 Contents Contents 1 1 Introduction 2 2 Background

More information

TELL ME MORE Step by Step ACTIVITY GUIDE

TELL ME MORE Step by Step ACTIVITY GUIDE TELL ME MORE Step by Step ACTIVITY GUIDE The following are the main components for each activity in this guide: ACTIVITY TITLE Activity's title as it appears in TELL ME MORE. Activity type and explanation

More information

AN EFFECTIVE METHOD FOR EDUCATION IN ACOUSTICS AND SPEECH SCIENCE Integrating textbooks, computer simulation and physical models

AN EFFECTIVE METHOD FOR EDUCATION IN ACOUSTICS AND SPEECH SCIENCE Integrating textbooks, computer simulation and physical models AN EFFECTIVE METHOD FOR EDUCATION IN ACOUSTICS AND SPEECH SCIENCE Integrating textbooks, computer simulation and physical models PACS: 43.10.Sv Arai, Takayuki Dept. of Electrical and Electronics Eng.,

More information

Phonetics-1 periodic frequency amplitude decibel aperiodic fundamental harmonics overtones power spectrum

Phonetics-1 periodic frequency amplitude decibel aperiodic fundamental harmonics overtones power spectrum 24.901 Phonetics-1 1. sound results from pressure fluctuations in a medium which displace the ear drum to stimulate the auditory nerve air is normal medium for speech air is elastic (cf. bicyle pump, plastic

More information

Level 1-Beginning ESL Literacy Program (CASAS 180 AND below)

Level 1-Beginning ESL Literacy Program (CASAS 180 AND below) For WASHINGTON STATE Adult LEARNING STANDARDS Level 1-Beginning ESL Literacy Program (CASAS 180 AND below) STANDARD I: Read With Understanding To read with understanding, Washington ESL students should:

More information

Critical Review: The Benefits of Auditory Training for Adults with Mild to Moderate Sensorineural Hearing Loss.

Critical Review: The Benefits of Auditory Training for Adults with Mild to Moderate Sensorineural Hearing Loss. Critical Review: The Benefits of Auditory Training for Adults with Mild to Moderate Sensorineural Hearing Loss. Gabriella Lesniak M.Cl.Sc AUD Candidate University of Western Ontario: School of Communication

More information

CHARTES D'ANGLAIS SOMMAIRE. CHARTE NIVEAU A1 Pages 2-4. CHARTE NIVEAU A2 Pages 5-7. CHARTE NIVEAU B1 Pages 8-10. CHARTE NIVEAU B2 Pages 11-14

CHARTES D'ANGLAIS SOMMAIRE. CHARTE NIVEAU A1 Pages 2-4. CHARTE NIVEAU A2 Pages 5-7. CHARTE NIVEAU B1 Pages 8-10. CHARTE NIVEAU B2 Pages 11-14 CHARTES D'ANGLAIS SOMMAIRE CHARTE NIVEAU A1 Pages 2-4 CHARTE NIVEAU A2 Pages 5-7 CHARTE NIVEAU B1 Pages 8-10 CHARTE NIVEAU B2 Pages 11-14 CHARTE NIVEAU C1 Pages 15-17 MAJ, le 11 juin 2014 A1 Skills-based

More information

Speech Synthesis by Artificial Neural Networks (AI / Speech processing / Signal processing)

Speech Synthesis by Artificial Neural Networks (AI / Speech processing / Signal processing) Speech Synthesis by Artificial Neural Networks (AI / Speech processing / Signal processing) Christos P. Yiakoumettis Department of Informatics University of Sussex, UK (Email: c.yiakoumettis@sussex.ac.uk)

More information

A Comparison of Speech Coding Algorithms ADPCM vs CELP. Shannon Wichman

A Comparison of Speech Coding Algorithms ADPCM vs CELP. Shannon Wichman A Comparison of Speech Coding Algorithms ADPCM vs CELP Shannon Wichman Department of Electrical Engineering The University of Texas at Dallas Fall 1999 December 8, 1999 1 Abstract Factors serving as constraints

More information

How can Lightspeed s sound panel solutions possibly break down acoustic barriers in the classroom?

How can Lightspeed s sound panel solutions possibly break down acoustic barriers in the classroom? How can Lightspeed s sound panel solutions possibly break down acoustic barriers in the classroom? There is no audio solution that throws sound. There is no solution that overcomes the laws of physics.

More information

Examining the Role of Talker-Specific Details in the Perception of Words Spoken by Famous Talkers

Examining the Role of Talker-Specific Details in the Perception of Words Spoken by Famous Talkers Cleveland State University EngagedScholarship@CSU ETD Archive 2009 Examining the Role of Talker-Specific Details in the Perception of Words Spoken by Famous Talkers Alisa M. Maibauer Cleveland State University

More information

Algorithm for Detection of Voice Signal Periodicity

Algorithm for Detection of Voice Signal Periodicity Algorithm for Detection of Voice Signal Periodicity Ovidiu Buza, Gavril Toderean, András Balogh Department of Communications, Technical University of Cluj- Napoca, Romania József Domokos Department of

More information

Audio Engineering Society. Convention Paper. Presented at the 129th Convention 2010 November 4 7 San Francisco, CA, USA

Audio Engineering Society. Convention Paper. Presented at the 129th Convention 2010 November 4 7 San Francisco, CA, USA Audio Engineering Society Convention Paper Presented at the 129th Convention 2010 November 4 7 San Francisco, CA, USA The papers at this Convention have been selected on the basis of a submitted abstract

More information

J. Acoust. Soc. Am. 112 (4), October /2002/112(4)/1675/6/$ Acoustical Society of America

J. Acoust. Soc. Am. 112 (4), October /2002/112(4)/1675/6/$ Acoustical Society of America Benefits of amplification for speech recognition in background noise Christopher W. Turner and Belinda A. Henry Department of Speech Pathology and Audiology, University of Iowa, Iowa City, Iowa 52242 Received

More information

Modeling Adaptation to a Novel Accent

Modeling Adaptation to a Novel Accent Modeling Adaptation to a Novel Accent Kasia Hitczenko 1 (khit@umd.edu) Naomi H. Feldman 1,2 (nhf@umd.edu) 1 Department of Linguistics and 2 Institute for Advanced Computer Studies 1401 Marie Mount Hall,

More information

The Icelandic Speech Recognition Project Hjal

The Icelandic Speech Recognition Project Hjal Eiríkur Rögnvaldsson University of Iceland The Icelandic Speech Recognition Project Hjal 1. The project In the end of 2002, the University of Iceland and four leading companies in the telecommunication

More information

Computer Networks and Internets, 5e Chapter 6 Information Sources and Signals. Introduction

Computer Networks and Internets, 5e Chapter 6 Information Sources and Signals. Introduction Computer Networks and Internets, 5e Chapter 6 Information Sources and Signals Modified from the lecture slides of Lami Kaya (LKaya@ieee.org) for use CECS 474, Fall 2008. 2009 Pearson Education Inc., Upper

More information

Prelinguistic vocal behaviors. Stage 1 (birth-1 month) Stage 2 (2-3 months) Stage 4 (7-9 months) Stage 3 (4-6 months)

Prelinguistic vocal behaviors. Stage 1 (birth-1 month) Stage 2 (2-3 months) Stage 4 (7-9 months) Stage 3 (4-6 months) 3 stages of phonological development 1. Prelinguistic vocal behaviors 2. Phonology of the first 50 words 3. Emergence of rules Prelinguistic vocal behaviors Reflexive/vegetative sounds (birth-1 month)

More information

Speech Production 2. Paper 9: Foundations of Speech Communication Lent Term: Week 4. Katharine Barden

Speech Production 2. Paper 9: Foundations of Speech Communication Lent Term: Week 4. Katharine Barden Speech Production 2 Paper 9: Foundations of Speech Communication Lent Term: Week 4 Katharine Barden Today s lecture Prosodic-segmental interdependencies Models of speech production Articulatory phonology

More information

The Impact of a Language Teaching Computer Game on ESL Students Recognition of Spoken and Written Word Forms

The Impact of a Language Teaching Computer Game on ESL Students Recognition of Spoken and Written Word Forms The Impact of a Language Teaching Computer Game on ESL Students Recognition of Spoken and Written Word Forms Amanda Müller Flinders University (Australia) amanda.muller@flinders.edu.au Abstract This study

More information

Areas of Processing Deficit and Their Link to Areas of Academic Achievement

Areas of Processing Deficit and Their Link to Areas of Academic Achievement Areas of Processing Deficit and Their Link to Areas of Academic Achievement Phonological Processing Model Wagner, R.K., Torgesen, J.K., & Rashotte, C.A. (1999). Comprehensive Test of Phonological Processing.

More information

Are Pitch Contour and Quantity Independent Distinctive Features in Bosnian Serbian?

Are Pitch Contour and Quantity Independent Distinctive Features in Bosnian Serbian? Are Pitch Contour and Quantity Independent Distinctive Features in Bosnian Serbian? IKP-Arbeitsbericht NF 14 Petra Wagner and Jelena Mandić pwa@ikp.uni-bonn.de This study investigates the independent phonological

More information

Talker-identification training using simulations of hybrid CI hearing : generalization to speech recognition and music perception

Talker-identification training using simulations of hybrid CI hearing : generalization to speech recognition and music perception University of Iowa Iowa Research Online Theses and Dissertations Summer 2014 Talker-identification training using simulations of hybrid CI hearing : generalization to speech recognition and music perception

More information

Word Learning Under Accent Variability in Monolinguals and Bilinguals Honors Project

Word Learning Under Accent Variability in Monolinguals and Bilinguals Honors Project Word Learning Under Accent Variability in Monolinguals and Bilinguals 2010-2011 Honors Project Kristin Muench Language Acquisition and Sound Recognition Lab Department of Cognitive Science, University

More information

The Linguistics, Phonology, and Phonetics Perspective

The Linguistics, Phonology, and Phonetics Perspective Oxford Scholarship Online You are looking at 1-10 of 12 items for: fulltext : science social point value major theories effect human linphon The Linguistics, Phonology, and Phonetics Perspective Mark Tatham

More information